Well, lately I’ve been tinkering around with antennas again for use whilst mobile. By mobile, I mean public transport mobile, and pedestrian mobile… I don’t own a car.
Most handheld radios come with rubber ducky antennas. My Kenwood TH-F7E is no exception. Rubber ducky “dummy loads” as I call them, are quite dismal performers against an efficient and resonant dipole — however they are usually much more convenient size-wise than a resonant antenna. It’s a similar story for mobile HF antennas… they’re a negative gain antenna.
Naturally, the way I’ve improved it, is to construct my own. So far, I’ve built a number of these antennas. The first version was a bit of stainless steel fencing wire with a SMA connector soldered to the botto and a few turns of a coil. It was about 50cm long… roughly cut, so possibly not that well tuned… but it performed quite well… allowing me to reach repeaters up to 80km away on 2m with minimal line-of-sight. The SMA->antenna interface proved to be a fragile component however.
A later version attempted to be a half-wave 2m, and could extend (fold out) to become a full-wave 70cm antenna. This again, had similar mechanical issues to its predicessor, but performed excellent otherwise.
It was pointed out to me that trying to match a full-wave antenna is asking a bit much of the power-amp in my radio, and that an odd multiple of 1/4 wavelength was better.
Thus, the third revision, I made using some surplus solid core electrical copper cable, and a BNC connector, was constructed at a length of 52cm. 52cm was calculated according to the rule v=f?… the frequencies I had in mind were 146.500MHz (2m FM calling frequency) and 439MHz (near to the input frequency of most 70cm repeaters). 52cm was calculated approximately as being 1/4 wavelength on 2m, and 3/4 wavelength on 70cm.
This antenna performed exceptionally well, and with the BNC connector, showed less mechanical problems than its predicessors. It did however, put a lot of strain on the BNC->SMA adaptors I was using, and I had to be extra careful with the antennas length.
Using MMANA-GAL, I tried modelling this antenna, just to see how well it infact should work in theory. Well I was right… about 6dBi gain on 70cm… it wasn’t bad at all! However, SWR was through the roof, 1:several hundred thousand.
I’ve since found a long forgotten RC car remote that operated on 40MHz, which had a 40cm long antenna. I’ve stretched the base coil out a bit to make it 42cm and added a SMA connector to the bottom of it… I may have to extend it a bit, but this seems like a closer match to what I’m looking for… but my modelling of it is just as dismal in terms of SWR.
Now, as far as improving SWR… the ways I know of to fix this problem are:
- To make the antenna a dipole, effectively doubling its length
- Adding a ground plane or counterpoise radial
- Adding a matching section (like a J-pole)
- Using some sort of matching network
The thing that has me curious, is the rubber ducky antenna. Now granted, I know those things do not radiate well. They must however be doing something to keep the finals from blowing. My understanding is that they’re little more than a spring in a plastic jacket… I can’t see how that is meant to match to the 50? source impedance.
The antenna design I use is apparently very similar to one used by Motorola for some of their professional radio antennas, according to this post. Now I’m not sure how 19.5″ is arrived at — it doesn’t fit with the maths I used above, there is something I’ve missed. Surely though, an antenna of this type would have the same impedance matching issues as the ones I’ve designed. Either there’s other magic involved, or the finals in many handheld radios are more hardy than I thought.
At some stage I might borrow a SWR meter or antenna analyser for VHF/UHF… it’d be interesting to see just how far off the mark I am. I haven’t blown my finals…yet. The radio seems to still have plenty of life in it. I’d be interested to know however, some of the background on this topic. There is something I am missing, and I can’t quite put my finger on it.